The proliferation and expansion of computer systems, networks, databases, the Internet, and particularly the World Wide Web (WWW), has resulted in a vast and diverse collection of information and means of communication. The current Internet infrastructure involves millions of computers linked together on a computer network. This network allows all of the computers to communicate with one another. Clients are typically linked to the Internet via Internet Service Providers (ISP's), which in turn connect to larger ISP's. This allows numerous clients to communicate to each other through their various connections.
In general, all the machines on the Internet can be categorized into two types: servers and clients. Typically, machines that provide services (like Web servers, FTP servers, Email servers, etc.) are servers. Servers are loaded with the appropriate software in order to allow them to perform their intended services. Machines that request information from servers are typically called clients. In order to differentiate between machines on the network, each machine is given a unique address called an IP address.
The IP address is a thirty-two bit number that is normally expressed as 4 octets in a dotted decimal number (e.g., 192.168.1.101). Each of the octets can have values between 0 and 255 (2 8 possibilities per octet). When a client connects to the Internet, the client is assigned an IP address through their Internet Service Provider (ISP) for the duration of the connection. Conversely, the IP addresses of servers are relatively static, and do not change very often.
Because it is difficult for clients to remember IP addresses, and because IP addresses need to change, most servers on the Internet possess domain names (e.g., “www.whoknowz.com”) to help users reach their intended servers without remembering strings of numbers. Name servers, used in the domain name system (DNS), map the human-readable names into IP addresses to help clients reach their destinations. When a client enters a domain name, the browser (via a resolver) extracts the domain name and passes it to a name server, which will return the correct IP address to the associated site. The Domain Name System is comprised of a distributed database and name servers that access that database.
One of the main problems with the current utilization of IP addresses and domain names on the World Wide Web (WWW) is that the WWW is based largely on the hypertext transport protocol (“HTTP-protocol”). The nature of HTTP protocol allows information, such as a client's e-mail address, web sites that were visited, and information on the client's software and host computer, to be recorded and traced by the server. This opens up the user to a range of privacy threats including unwanted e-mails, solicitations, and “cookies” (data that is stored on the client's machine by a server and subsequently used for identification).
Furthermore, clients that wish to cloak themselves from such intrusions are forced into systems that simply provide alternate account identities for the client; while the client is protected, the alternate account identity becomes the object of the unwanted e-mails, “cookies”, etc., instead. The effect of this is similar to the client manually creating a new user account in which to browse the WWW.
One of the solutions available is to route the client through a proxy server in order to substitute IP information being sent by the client. When a client desires to visit a web server, the packets sent from client's computer are routed through a proxy server. At the proxy server, the server executes algorithms to extract information that would identify the client, and replaces the information with predetermined substitutes.
Afterwards, the proxy server routes the packet out to the web server. Once the web server receives the packet, all of the information points back to the proxy server, and not to the client. This in effect “hides” the client from the web server.
However, a drawback to such systems is that, as mentioned before, the client is obtaining protection merely through the use of an alternate identity that is ultimately assigned back to the same client. Furthermore, current systems do not have any added flexibility designed in the system to take advantage of anonymous client group browsing or multiple group association. In order to fully take advantage of ad hoc identity browsing, additional features need to be added in order to create a “community-like” environment among numerous anonymous clients.